RobustSplat: Decoupling Densification and Dynamics for Transient-Free 3DGS

TL;DR

RobustSplat introduces a novel approach to 3D Gaussian Splatting that effectively reduces artifacts caused by transient objects, enhancing scene rendering accuracy through delayed densification and multi-resolution mask bootstrapping.

Contribution

The paper proposes two key innovations: a delayed Gaussian growth strategy and a scale-cascaded mask bootstrapping method, improving transient object handling in 3D scene rendering.

Findings

Outperforms existing methods on multiple datasets

Reduces artifacts caused by transient objects

Enhances robustness and scene detail accuracy

Abstract

3D Gaussian Splatting (3DGS) has gained significant attention for its real-time, photo-realistic rendering in novel-view synthesis and 3D modeling. However, existing methods struggle with accurately modeling scenes affected by transient objects, leading to artifacts in the rendered images. We identify that the Gaussian densification process, while enhancing scene detail capture, unintentionally contributes to these artifacts by growing additional Gaussians that model transient disturbances. To address this, we propose RobustSplat, a robust solution based on two critical designs. First, we introduce a delayed Gaussian growth strategy that prioritizes optimizing static scene structure before allowing Gaussian splitting/cloning, mitigating overfitting to transient objects in early optimization. Second, we design a scale-cascaded mask bootstrapping approach that first leverages lower-resolution feature similarity supervision for reliable initial transient mask estimation, taking advantage of its stronger semantic consistency and robustness to noise, and then progresses to high-resolution supervision to achieve more precise mask prediction. Extensive experiments on multiple challenging datasets show that our method outperforms existing methods, clearly demonstrating the robustness and effectiveness of our method. Our project page is https://fcyycf.github.io/RobustSplat/.